Full Article: PDF
Scientific Object Identifier: http://s-o-i.org/1.1/TAS-03-59-1
DOI: https://dx.doi.org/10.15863/TAS.2018.03.59.1
Language: Russian
Citation: Korneev AM, Sukhanov AV, Shipulin IA (2018) OPTIONS OF ORGANIZATION AND GENERAL SCHEMES OF THE STOCHASTIC OPTIMIZATION ALGORITHM BASED ON THE METHOD OF SIMULATING ANNEALING. ISJ Theoretical & Applied Science, 03 (59): 1-5. Soi: http://s-o-i.org/1.1/TAS-03-59-1 Doi: https://dx.doi.org/10.15863/TAS.2018.03.59.1 |
Pages: 1-5
Published: 30.03.2018
Abstract: The article gives an overview of the existing modifications of the optimization algorithm simulated annealing (simulated annealing) based on the Markov random search in discrete multidimensional spaces. The theoretical foundations for constructing schemes for the stochastic search for a global minimum of the objective function f: Rn by the annealing simulation method are formulated. The principles of influence of numerical parameters are described - temperature, energy of the system, boundaries of the search area, the number of iterations, - the rate of convergence of the algorithm and the accuracy of the values determined with its help. The description of three modifications of the random search algorithm simulated annealing (modifications A, B, C) is given, their advantages and disadvantages are separately highlighted. A block diagram of the complete procedure for finding the minimum of the objective function by the annealing simulation method is presented. A comparative analysis of the Boltzmann scheme and the Cauchy scheme of the organization of the random search algorithm of the global minimum simulated annealing is carried out. Examples of the use of modifications of the random search algorithm simulated annealing in practice are given.
Key words: random search, annealing simulation algorithm, random variable, stop parameter, chemical composition, composite structure, cast iron expansion coefficient, specific heat of cast iron.
|
